Markerless registration approach using dynamic touchable region model

Abstract

Background: Markerless registration is required for image-guided surgery; it has limited accuracy due to the ambiguity of the correspondence point set. In this study, we proposed a registration framework to improve registration accuracy for markerless registration using a dynamic touchable region model (DTRM).

Methods: The DTRM is defined using the geometric characteristics of the surface around the fiducial area using Intrinsic Shape Signature keypoints. The new registration procedure, which combines the DTRM with iterative closest point-based registration (ICDTP) was implemented and verified with phantom experiments for a single-user and multi-user study.

Results: The ICDTP registration framework provides improved performance over the paired-point registration for surgeries, where it is inappropriate to construct the corresponding positions with adhesive fiducial markers.

Conclusion: The proposed method can be an alternative approach for image-guided surgery in operations for which itis not appropriate to set up markers for registration.